<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1427">
  <Title>Robust, Applied Morphological Generation .... ......... ..... . _</Title>
  <Section position="6" start_page="206" end_page="206" type="concl">
    <SectionTitle>
5 Conclusions provided to us by the University of Sheffield
</SectionTitle>
    <Paragraph position="0"> We have described a generatorf0r English in:: ' G.A~E ~projoet-,...:(3hris -Brew,,.Dale- Gerdem:an.n~..</Paragraph>
    <Paragraph position="1"> flectional morphology. The main features of the Adam Kilgarriff and Ehud Reiter have suggenerator are: wide coverage and high accuracy It incorporates data from several large corpora and machine readable dictionaries. An evaluation has shown the error rate to be very low.</Paragraph>
    <Paragraph position="2"> robustness The generator does not contain an explicit lexicon or word-list, but instead comprises a set of morphological generalisations together with a list of exceptions for specific (irregular) words. Unknown words are very often handled correctly by the generalisations. null maintainability and ease of use The organisation into generalisations and exceptions can save development time since addition of new vocabulary that has regular morphology does not require any changes to be made. The generator is packaged up as a Unix filter, making it easy to integrate into applications.</Paragraph>
    <Paragraph position="3"> speed and portability The generator is based on efficient finite-state techniques, and implemented using the widely available Unix Flex utility.</Paragraph>
    <Paragraph position="4"> freely available The morphological generator and the orthographic postprocessor are fi'eely available to the NLG research community. See &lt;http://www.cogs.</Paragraph>
    <Paragraph position="5"> susx.ac.uk/lab/nlp/carroll/morph.html&gt;.</Paragraph>
    <Paragraph position="6"> In future work we intend to investigate the use of phonological information in machine readable dictionaries for a more principled solution to the consonant doubling problem. We also plan to further increase the flexibility of the generator by including an option that allows the user to choose whether it has a prei~rence for generating British or American English spelling.</Paragraph>
  </Section>
class="xml-element"></Paper>